Algorithms and rendering

When you stretch an audio clip, AudioSnap uses a particular formula, or algorithm, to stretch the audio. The best algorithms take the most computing time and power to complete, so if AudioSnap always used the best available algorithms (the iZotope Radius algorithms), you would not be able to play back your clip in a reasonable amount of time to listen to the timing. That’s why you can choose a quick algorithm for “online” rendering (real-time playback), and a different algorithm for “offline” or non-real-time rendering. Offline rendering in an AudioSnap context usually means bouncing to track, but it also refers to other offline processes such as freezing tracks and applying effects. When you do any of these mixdown operations such as bouncing, exporting, or freezing, audio clips use the offline rendering algorithm you have chosen for them.

Note: The Online stretch method is for preview purposes only during playback. The final audio quality will be greatly improved after the Offline stretch method is applied during mixdown/export.

Typical algorithm choices for an AudioSnap session work like this:

Do your AudioSnap time stretching, error correction, etc., with a quick algorithm: choose either Percussion or GrooveClip.

After your AudioSnap editing is finished, bounce to track with a better algorithm: choose one of the Radius choices, or for drum tracks, Percussion is usually the best choice.

Note: You should avoid rendering AudioSnap clips more than once. Applying stretch algorithms multiple times in succession can degrade the audio quality (similar to transcoding an MP3 file). Until your project is mixed and finalized, it is recommended that you use the Freeze function instead of Bounce to Track(s) or Bounce to Clip(s) if you need to temporarily off-load CPU processing power. For details, see To freeze an AudioSnap-enabled clip.

The following table describes the available algorithms.

Algorithm	Description
Groove	Good all-around algorithm to use while stretching audio
Percussion	Gives a better sound on percussion clips while stretching audio
iZotope Radius Mix	Best all-around choice when bouncing to track with audio clips that contain a mix of different sounds
iZotope Radius Mix Advanced	This choice is similar to iZotope Radius Mix, but exposes two parameters: Pitch coherence. Pitch coherence preserves naturalness of timbre for pitched solo voices, such as human speech, saxophone or vocals. The default value for this control is 50%. Higher values of pitch coherence will increase phase accuracy but may sound more modulated. Phase coherence. When processing stereo or surround audio, Phase coherence helps to preserve the phase relationships between channels. The default value for this control is 50%. Higher values of phase coherence will preserve the phase between multiple audio channels at the expense of slight distortion. The Pitch coherence and Phase coherence parameters can be adjusted in Edit > Preferences > Audio - Playback and Recording.
iZotope Radius Solo	Best choice when bouncing to track with audio clips of a solo instrument. Although a guitar is a “solo” instrument, if you play chords instead of single notes you may want to select “Radius Mix” instead. Tip: if a clip/track contains a mixture of polyphonic and monophonic parts, you may want to split the parts into separate clips and experiment with different algorithms for each clip.
iZotope Radius Solo (Bass)	Best choice when bouncing to track with audio clips of a bass instrument
iZotope Radius Solo (Vocal)	Best choice when bouncing to track with audio clips of a solo voice
Elastique Pro	Elastique Pro is a general purpose high quality time-stretching engine that fulfills the demands of professional productions and broadcast applications. It minimizes stretching artifacts, offers stable timing, inter-channel phase coherence and sample accurate stretching which allows for sharp transients and crystal clear vocals.
Elastique Efficient	Elastique Efficient gives you similar time-stretching quality as Elastique Pro, but with a lower CPU hit. The algorithm is targeted at complex polyphonic signals like complete mixes and offers the same transient preservation as Elastique Pro.

To specify default stretch methods

You can specify the default global online and offline stretch methods in the Preferences dialog box, for both regular audio clips and VocalSync clips.

The stretch method determines which algorithm Sonar uses for AudioSnap, stretching, pitch shifting, and bouncing audio clips.

1.	Go to Edit > Preferences > Audio - Playback and Recording.

2.	Under Stretch Methods, select the desired default Online Render and Offline Render modes. The online render mode is used for real-time playback, and the offline render mode is used for export, freeze, and bounce.

Sonar uses the default stretch methods whenever the Online Render or Offline Render mode is set to Default in the Track Properties Inspector or Clip Properties Inspector (the inherited stretch method name is shown in parenthesis), but you can override the default render modes at the track and clip level.

The following diagram shows how the default stretch methods propagate to tracks and clips.

 

Note: You can also specify the global, per track, and per clip stretch methods in the AudioSnap palette. For details, see Using the AudioSnap palette.

To choose default algorithms

1.	Press ALT+A to open the AudioSnap palette.

2.	In the Applies To list, select Global.

3.	Under Stretch Methods, select the desired Online and Offline stretch methods.

Every AudioSnap-enabled clip that you stretch or bounce uses the settings you choose for default stretch algorithms, except clips that you choose individual settings for.

To choose render algorithms for an individual clip

1.	Select the clip or clips you want to edit.

2.	Press ALT+A to open the AudioSnap palette.

The AudioSnap palette displays the current stretch method of the selected clip. If you select multiple clips that have different stretch methods, the Online and Offline lists display (Multi).

3.	In the Applies To list, select Clips.

4.	Under Stretch Methods, select the desired Online and Offline stretch methods.

To choose render algorithms for all current and future clips on a track

1.	Select the track or tracks you want to edit.

2.	Press ALT+A to open the AudioSnap palette.

3.	In the Applies To list, select Tracks.

4.	Under Stretch Methods, select the desired Online and Offline stretch methods.

All current and future clips in the selected tracks will inherit the selected stretch methods.

To freeze an AudioSnap-enabled clip

1.	Do one of the following to open the Freeze Options dialog:

In the Track view, right-click any track’s Freeze button .

Right-click an audio track, and select Freeze > Freeze Options from the pop-up menu.

2.	Make sure the Track FX check box is cleared.

3.	Click OK to close the Freeze Options dialog box.

4.	Do one of the following:

In the Track view, click the desired track’s Freeze button .

Right-click the track and select Freeze > Freeze Track from the pop-up menu.

Sonar bounces the audio in the track to a new audio clip or clips.

To bounce an AudioSnap-enabled clip

1.	Select the clip you want to bounce.

2.	Click the Track view Tracks menu and choose Bounce to Track(s) to open the Bounce to Track(s) dialog box.

3.	In the Destination field, choose the track where you want your new audio clip to go.

4.	In the Source Category field, choose Tracks.

5.	Choose the Channel Format and Dithering options you want.

6.	Enable all the options in the Mix Enables section.

7.

Click OK.

 

See:

Freeze tracks and synths

To bounce multiple audio clips to a new track

To bounce to clips

Using global tools on AudioSnap transient markers

Editing transient markers

Using the AudioSnap palette

Synchronizing audio and the project tempo

Fixing timing problems in audio clips

Extracting MIDI timing information from audio

General editing

Using the Pool